Sentiment Analysis on Cryptocurrencies

“Investors on ‘suicide watch’ after disastrous plunge in value of cryptocurrencies Bitcoin and Ethereum” 
(https://www.thesun.co.uk/money/7026138/cryptocurrency-bitcoin-crash-suicide-fears/)

“‘They sense panic’: Bitcoin, Ether plunge as selloff grips cryptocurrency market” 
(https://business.financialpost.com/technology/blockchain/they-sense-panic-bitcoin-ether-plunge-as-selloff-grips-cryptocurrency-market)

“Bitcoin, Ether Sink as ‘Sense of Panic’ Grips Crypto Investors” 
(https://www.bloomberg.com/news/articles/2018-08-14/bitcoin-sinks-below-6-000-as-almost-everything-crypto-tumbles)


Take a look at the above headlines that were dominating the crypto news cycle just days ago. It has amazed me for some time how emotional the crypto markets can be. Amidst this sea of negativity and fearfulness, I could not help but think that this was indeed a moment to be greedy, à la Warren Buffet. It is tough though, to figure out at any particular moment whether the markets are indeed fearful or greedy. In the world of finance, Sentiment Analysis appears to be our best bet in doing so.

Sentiment Analysis is a concept that has been around for some time and has had applications in the world of marketing and customer service for some time. The basic premise is to determine a speaker’s attitude vis-à-vis a particular topic. (https://en.wikipedia.org/wiki/Sentiment_analysis)
For finance, the idea is to come up a more sophisticated way of determining whether the market views a particular asset positively or negatively. Traders can then profit by taking a contrarian view if it becomes clear that the market is overly positive or negative on the particular asset. 
(https://www.investopedia.com/articles/01/030701.asp)

Several traditional tools have been developed for the finance world, but in the crypto world, there are also a host of resources one can take advantage of. There is a particular article on hackernoon (https://hackernoon.com/@bigurb) that I cannot recommend enough. The writer, Blake Urban, details his strategy for mixing Technical and Sentiment Analysis in crypto-trading, but more importantly, sums up the best resources to gauge crypto market sentiment.

I would suggest spending some time reading the article, as well as reading up on the different tools suggested. For instance, my favorite tool is the Crypto Fear and Greed Index, not only because it is easy to read, but because they explain with great clarity the different factors used to compile the index (social media, surveys, market volatility, etc.).


Beyond the tools, I have become increasingly convinced that Sentiment Analysis has an outsized importance in crypto-trading; even more so than it does for trading equities or other typical commodities. This is so for a number of reasons:

  1. Digital communication — The hyper-connectivity of the crypto world, through news sites and social media platforms like Telegram, exacerbates fear and greed in a way that Sentiment Analysis can take advantage of. It is not a stretch to imagine that that crypto investors and traders typically communicate and consume content on digital channels (more so than the average person), which means they are communicating at an extraordinarily high frequency. As such, this can create vicious cycles of emotion where everyone feeds of each other’s undue pessimism or unjustified exuberance at an incredible pace.
  2. Difficulties with Fundamental Analysis — There are a number of things investors consider when doing their fundamental analysis on the various cryptos, but the fact remains that cryptocurrencies are still a relatively new asset without significant literature to guide valuation and assessment. Academics and professionals have devoted decades to studying the more traditional assets, something crypto does not benefit from. This inevitably makes relying on Fundamental Analysis a lot more tenuous, which shifts the emphasis to Technical or Sentiment Analysis. In fact, unlike Fundamental Analysis, much of the literature on TA or SA can map to cryptocurrencies as these approaches generally give less consideration to the nature of the asset in question.
  3. Usefulness of Tools — This harks back to the fact that much of the crypto world communicates over the internet. There is so much digital ink on Twitter, Reddit, Telegram, etc. specifically regarding cryptocurrencies that the SA tools have plenty of data to take advantage of, thus improving their effectiveness.

Leave a Reply

Your email address will not be published. Required fields are marked *